- Email: [email protected]
Meta-analysis of long-term outcomes of drug-eluting stent implantations for chronic total coronary occlusions
Meta-analysis of long-term outcomes of drug-eluting stent implantations for chronic total coronary occlusions Jian Ma, MD, PhDa,y, Weiwei Yang, MDb,y, Manpreet Singh c, Tianqing Peng, MDd, Ningyuan Fang, MD, PhDb, Meng Wei, MD, PhDa,* a
Division of Cardiology, Shanghai Sixth People’s Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China b Division of Geriatrics, Shanghai Ren-Ji Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China c Department of Pathology, University of Western Ontario, London, Ontario, Canada d Critical Illness Research, Lawson Health Research Institute, Department of Medicine and Department of Pathology, University of Western Ontario, London, Ontario, Canada
article info
abstract
Article history: Received 21 May 2009 Revised 4 July 2010 Accepted 17 July 2010 Online 21 March 2011
Background: In the treatment of chronic total occlusions (CTOs), some uncertainty exists regarding the effect of drug-eluting stents (DESs) compared with the effects of bare mental stents (BMSs). We reviewed outcomes of DES vs. BMS implantation for CTO lesions, to evaluate the risk-benefit ratio of DES implantation.
Keywords: Drug-eluting stents Bare mental stents Chronic total occlusions Meta-analysis
Methods: Relevant studies of long-term clinical outcomes or angiographic outcomes of both BMS and DES implantation were examined. The primary endpoint comprised major adverse cardiovascular events (MACEs), including allcause deaths, myocardial infarctions (MIs), and target lesion revascularizations (TLRs). A fixed-effect model and random-effect model were used to analyze the pooling results. Results: Ten studies were included according to the selection criteria. Eight were nonrandomized controlled trials, and two consisted of a randomized controlled comparison between DES and BMS implantation. No significant difference was evident for in-hospital MACE rates between the two groups (odds ratio [OR], 1.07; 95% confidence interval [CI], .53 to 2.13), but the long-term MACE rates in the DES group were significantly lower than in the BMS group (OR, .22; 95% CI, .13 to .38; P < .00001). The rates of stent restenosis and reocclusions were also significantly lower in the DES group (OR, .14; 95% CI, .09 to .20; and OR, .23; 95% CI, .12 to .41, respectively). Conclusion: Implantation of the DES improves long-term angiographic and clinical outcomes compared with BMS in the treatment of CTO lesions. Cite this article: Ma, J., Yang, W., Singh, M., Peng, T., Fang, N., & Wei, M. (2011, MAY/JUNE). Meta-analysis of long-term outcomes of drug-eluting stent implantations for chronic total coronary occlusions. Heart & Lung, 40(3), e32-e40. doi:10.1016/j.hrtlng.2010.07.009.
* Corresponding author: Meng Wei, MD, PhD, Division of Cardiology, Shanghai Sixth People’s Hospital, No. 600 Yishan Road, Shanghai 200233, People’s Republic of China. E-mail address: [email protected] (M. Wei). y These authors equally contributed to this work 0147-9563/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.hrtlng.2010.07.009
h e a r t & l u n g 4 0 ( 2 0 1 1 ) e 3 2 ee 4 0
A chronic total coronary artery occlusion (CTO) is defined as the complete obstruction of coronary arteries with thrombolysis in myocardial infarction (TIMI) flow grade 0, with an estimated duration of >3 months. It is one of the most challenging lesion subsets in percutaneous coronary intervention (PCI). Although bare mental stents (BMSs) are widely applied clinically and, upon successful recanalization of CTO, were demonstrated to improve left ventricular function and survival,1-3 the major complications of high rates of restenosis and reocclusions limit its long-term usefulness.4,5 Drug-eluting stents (DESs), including sirolimus-eluting and paclitaxel-eluting varieties (SES and PES, respectively), are able to reduce the risk of restenosis.6,7 A few studies reported, with conflicting results, the clinical and angiographic outcomes of DESbased PCI for the treatment of CTO compared with BMS-based PCI. However, most of these were limited because they were nonrandomized controlled trials (RCTs) with a small sample size. The aim of our metaanalysis was to evaluate the risk-benefit balance of DES implantation vs. that of BMS implantation in the treatment of CTO.
Methods Search Strategy and Selection Criteria Two trained investigators (J.M. and W.Y.) independently searched Medline, EMbase, the Chinese Biomedical Database (from 1980 to April 2009), and the Cochrane Library (Issue 3, 2009). The reference lists of retrieved papers and reviews were also checked. No language restriction was used. The main search keywords included “chronic total occlusions,” “CTO,” “drugeluting stents,” “DES,” and “clinical trial*.” The inclusion criteria for retrieved studies comprised: (1) comparisons of DES implantation vs. BMS implantation, (2) patients with CTO diseases, and (3) a follow-up period of at least 6 months. Ongoing studies, duplicate publications, and those with incomplete data or a lack of baseline or follow-up data were excluded from this analysis.
Data Extraction and Study Characteristics Two investigators (J.M. and W.Y.) independently selected trials and extracted data on prespecified data forms, with disagreements resolved by discussion between them or, when necessary, in consultation with an arbiter (M.W). The primary endpoint comprised major adverse cardiovascular events (MACEs). The secondary end points included all-cause death, myocardial infarction (MI), target vessel revascularization (TVR), target lesion revascularization (TLR), target vessel failure (TVF), and * The asterisk is a truncated symbol, indicating to search this word by right truncation.
e33
angiographic outcomes involving minimal lumen diameter (MLD), late lumen loss, restenosis, and reocclusion. The MACEs were a composite of all-cause death, MIs, and TLRs. A TLR was defined as an ischemiadriven percutaneous or surgical revascularization of the target lesion attributable to restenosis within the stent, or 5 mm distal or proximal to the stent, after the initial procedure. A TVR was defined as a repeated revascularization within the treated vessel. Target vessel failures were defined as a composite of deaths from cardiac causes, MIs, and ischemia-driven TVRs. Late lumen loss was determined as the change in MLD from immediately after stent implantation to followup. Restenosis was defined as a diameter stenosis of 50% both within the stent and in the adjacent segments 5 mm proximal and distal, as determined by follow-up angiography. Reocclusion was defined as a recurrent total occlusion at the previous angioplasty site.
Quality Assessment Two investigators (W.Y. and J.M) independently assessed the quality of eligible articles according to the established methods of the Cochrane Handbook,8 which includes 5 items: randomization, allocation concealment, blinding, reported loss to follow-up, or withdrawal and comparability of baseline. Disagreements were resolved by discussions between them and, if necessary, by consultation with an arbiter (M.W.).
Statistical Analysis Statistical analysis was performed using Review Manager 5.0.18 (Cochrane Collaboration, Oxford, UK). Dichotomous variables were reported as odds ratios (ORs) and 95% confidence interval (CIs), whereas continuous variables were reported as weighted mean differences (WMDs) and 95% CIs. The c2 test and I2 values were calculated as a measure of statistical heterogeneity.8 If the I2 value was 0%, it denoted no observed statistical heterogeneity among studies, whereas a greater I2 value denoted more increasing heterogeneity.9 If no statistical heterogeneity was evident, dichotomous and continuous variables were compared using a fixed-effects model. If statistical heterogeneity was evident among studies, we first found the potential reasons, and then performed a sensitivity analysis or subgroup analysis, or analyzed studies using a random-effects model. A funnel plot was performed to assess the presence of publication bias.
Results The initial search generated 243 reports, of which 233 were excluded because of inconsistencies with
e34
Table 1 e Characteristics of 10 included studies Study
DM (%)
LVEF (%)
HBP (%)
Previous MI (%)
Hyperlipidemia (%)
96
64 9
77
31
68 16
79
45
76
AþC
Hoye et al (2004)11 Lei et al (2005)12
84 381
60 11 61 10
76 87
12 22
UN 53 11
39 45
52 61
56 49
AþC A þ T/C
Nakamura et al (2005)13 Kelbaek et al (2006)14 Suttorp (2006)15
180
69 8
65
33
54 7
59
36
34
AþC
127 200
62 10 60 10
79 80
20 13
51 11 UN
41 46
60 49
89 90
AþC AþC
Werner et al (2006)16
122
63 10
78
33
64 17
77
45
81
AþC
Migliorini et al (2006)17 Hecto et al (2007)18
118
66 11
UN
19
UN
53
50
47
AþC
147
61 11
71
10
UN
39
51
62
AþC
Felice et al (2008)19
223
62 10
76
31
51 10
71
67
66
A þ T/C
Werner et al (2004)10
Oral antiplatelet regimen
Duration of antiplatelet regimen DES 6 m; BMS 1m 6m T/C 3 m (DES), 1 m (BMS), A maintained UN 12 m C 6 m; A indefinite DES 6 m; BMS 1 m C 6 m; A indefinite A þ C 6 m; A lifelong T/C 6 m (DES), 1 m (BMS); A indefinite
SES/ PES (%)
Clinical followup
Angiographic follow-up
0/100
12 m
6m
100/0 100/0
12 m 6m
6m 6-8 m
100/0
12 m
6-12 m
100/0 100/0
7m 6-12 m
6m 6m
0/100
12 m
6m
UN
6m
6m
100/0
36 m
36 m
73/38
18 m
18 m
DM, diabetes mellitus; LVEF, left-ventricular ejection fraction; HBP, high blood pressure; MI, myocardial infarction; DES, drug-eluting stents; BMS, bare mental stents; SES, sirolimus-eluting stents; PES, paclitaxel-eluting stents; A, aspirin; C, clopidogrel; T, ticlopidine; UN, unclear; m, months; y, years.
h e a r t & l u n g 4 0 ( 2 0 1 1 ) e 3 2 ee 4 0
No. of Mean Males patients age (y) (%)
h e a r t & l u n g 4 0 ( 2 0 1 1 ) e 3 2 ee 4 0
Figure 1 e Funnel plot of drug-eluting stents vs. bare mental stents for chronic total occlusions.
selection criteria. Ultimately, 10 published studies were included,10-19 enrolling 1678 subjects undergoing a PCI with DES or BMS implantations. Two studies14,15 were RCTs, and 810-13,16-19 were nonrandomized controlled comparisons. All studies were published between 2004 and 2008 inclusive. Among these, 6 studies11-15,18 used SES for the treatment of CTOs, and 2 studies10,16 used PES for the treatment of CTOs, whereas the remainder17,19 used both SES and PES for the treatment of CTOs. The characteristics of the included studies are presented in Table 1. The mean age of patients was 63 years. On average, males accounted for 76.5% of subjects. The clinical and angiographic median follow-up periods were 13 months (range, 6 to 36 months) and 6 months (range, 6 to 12 months), respectively. The proportion of patients with diabetes mellitus ranged from 10% to 33% across trials. About 52% of the patients had experienced a previous MI. Approximately 55% and 65% of patients had a history of hypertension and hyperlipidemia, respectively. The average baseline left-ventricular ejection fraction (LVEF), as reported in six studies,10,12-14,16,19 was 57% 12% standard deviation (SD). Funnel plots did not suggest any evidence of publication bias (Figure 1).
Data Synthesis With respect to all endpoints, a fixed-effects model was used to analyze statistical heterogeneity among studies, and to combine data from enrolled studies. The results are shown in Figures 2 to 6 and Table 2.
e35
group (Figure 2). With regard to long-term MACE rates, however, significant statistical heterogeneity was evident among studies (P ¼ .002; I2, 65%). Inconsistent MACE definitions in different studies could have created the statistical heterogeneity. The different clinical settings in which procedures were performed also added heterogeneity to the analysis. Moreover, the number of enrolled patients varied widely among studies and between the 2 groups, which also made long-term MACE rates different. When reanalyzed using the random-effects model, the long-term MACE rates were significantly lower in the DES group than in the BMS group (OR, .22; 95% CI, .13 to .38; P < .00001) (Figure 3). To eliminate the statistical heterogeneity, we performed a subgroup analysis of the primary endpoint according to different follow-up periods. In studies limiting follow-up periods to 6 months,12,15,17 no statistical heterogeneity was evident (P ¼ .57; I2, 0%), but the long-term MACE rates were significantly lower in the DES-based PCI group than in the BMSbased PCI group (OR, .35; 95% CI, .23 to .54; P < .00001). In studies extending follow-up periods to >6 months,10,11,13-16,18,19 statistical heterogeneity was still present (P ¼ .003; I2, 68%). Finally, sensitivity analyses were conducted after the exclusion of 2 studies.11,18 No statistical heterogeneity was evident among studies, and the DES group had significantly lower long-term MACE rates than the BMS group (OR, .13; 95% CI, .09 to .20; P < .00001).
MI Data on the rate of in-hospital and long-term MIs were available in 5 studies10,12,13,16,19 and 7 studies,11-15, 18, 19 respectively. As depicted in Figures 4 and 5, no statistical heterogeneity was evident among studies (P ¼ .89; I2, 0%; and P ¼ .94; I2, 0%, respectively), with no significant difference between the DES and BMS groups (OR, 1.14; 95% CI, .57 to 2.29; P ¼ .72; and OR, .84; 95% CI, .46 to 1.54; P ¼ .58, respectively) (Figures 4, 5).
All-cause Death The rate of long-term mortality, including cardiac and noncardiac death, was reported in 7 studies.10,12,15-19 As shown in Figure 6, no statistical heterogeneity was evident among studies (P ¼ .80; I2, 0%), with no significant difference between the DES and BMS groups (OR, 1.02; 95% CI, .49 to 2.10; P ¼ .97) (Figure 6).
TLR
Clinical Endpoints MACE Five studies10,12,13,16,19 reported both in-hospital and long-term rates of MACE, whereas the remainder11,14,15,17,18 only described long-term MACE rates. A subanalysis was performed according to different observation times. Regarding in-hospital MACE rates, no statistical heterogeneity (P ¼ .93; I2, 0%) and no significant difference (OR, 1.07; 95% CI, .53 to 2.13; P ¼ .86) were present between the DES group and BMS
Data on long-term TLR were available in 8 studies.10,12-16,18,19 Statistical heterogeneity was evident among these studies (P ¼ .05; I2, 51%). Sensitivity analyses, conducted after the exclusion of 1 study,18 showed no statistical heterogeneity (P ¼ .29; I2, 18%) but a significantly lower rate of TLR in the DES group vs. the BMS group (OR, .13; 95% CI, .08 to .19, P < .00001).
TVR Regarding the endpoint of long-term TVR, data were available in 7 studies.11,12,14,15,17-19 No statistical
e36
h e a r t & l u n g 4 0 ( 2 0 1 1 ) e 3 2 ee 4 0
Figure 2 e Pooled results of in-hospital major adverse cardiovascular events in patients undergoing percutaneous drug-eluting stent (DES) placement for chronic total occlusions. BMS, bare mental stents, M-H, Mantel-Haenszel.
Figure 3 e Pooled results of long-term MACEs in patients undergoing percutaneous drug-eluting stent (DES) placement for chronic total occlusions. BMS, bare mental stents, M-H, Mantel-Haenszel.
Figure 4 e Pooled results of in-hospital myocardial infarctions in patients undergoing percutaneous drugeluting stent (DES) placement for chronic total occlusions. BMS, bare mental stents, M-H, Mantel-Haenszel.
heterogeneity was evident among studies (P ¼ .29; I2, 18%), but the long-term TVR rate in the DES group was significantly lower than in the BMS group (OR, .25, 95% CI, .17 to .36; P < .00001).
TVF Three studies10,14,15 reported the long-term rates of TVF. No statistical heterogeneity (P ¼ .32; I2, 11%) was found among studies. Compared with the BMS group, the rate of long-term TVF was significantly lower in the DES group (OR, .16; 95% CI, .09 to .29; P < .00001).
Angiographic Outcomes The long-term coronary angiography analysis comprised 10 studies of 479 patients in the DES group, and 604 patients in the BMS group. The average angiographic follow-up time was 6 months. With regard to all angiographic endpoints except for late lumen loss, no statistical heterogeneity was evident among studies. Compared with the BMS group, the rate of MLD was significantly higher in the DES group (WMD, 1.00; 95% CI, .91 to 1.09; P < .00001), but the rate
e37
h e a r t & l u n g 4 0 ( 2 0 1 1 ) e 3 2 ee 4 0
Figure 5 e Pooled results of long-term myocardial infarctions in patients undergoing percutaneous drugeluting stent (DES) placement for chronic total occlusions. BMS, bare mental stents, M-H, Mantel-Haenszel.
Figure 6 e Pooled results of long-term death due to any cause in patients undergoing percutaneous drugeluting stent (DES) placement for chronic total occlusions. BMS, bare mental stents, M-H, Mantel-Haenszel.
Table 2 e Summary effects for subanalysis of MACEs, TLR, TVR, TVF, and angiographic endpoints Outcomes Subanalysis of MACEs Limiting follow-up to 6 months Extending follow-up to >6 months TLR TVR TVF Angiographic endpoints MLD Late lumen loss Restenosis Reocclusion
OR (95% CI) .35 .13 .13 .25 .16
WMD (95% CI)
(.23 to .54) (.09 to .20) (.08 to .19) (.17 to .36) (.09 to .29) 1.00 (.91 to 1.09) 1.04 (1.21 to .07)
.14 (.09 to .20) .23 (.12 to .41)
P
I2 (%)
.57 .11 .29 .29 .32
0 45.0 17.8 18.0 11.4
.17 .07 .24 .43
34.9 53.2 24.8 0
MACEs, major adverse cardiovascular events; TLR, target lesion revascularization; TVR, target vessel revascularization; TVF, target vessel failure; MLD, minimal lumen diameter; OR, risk ratio; WMD, weighted mean difference; 95% CI, 95% confidence interval.
of late lumen loss was significantly lower in the DES group (WMD, 1.04; 95% CI, 1.21 to .87; P < .00001). Furthermore, the rates of stent restenosis and reocclusion in the DES group were significantly lower than in the BMS group (WMD, .14; 95% CI, .09 to .20; P < .00001; and WMD, .23; 95% CI, .12 to .41, P < .00001, respectively).
Discussion The present meta-analysis focused on the clinical and angiographic efficacy of DES implantation for the treatment of CTO lesions compared with BMS implantation. The results indicate that DES (either SES
e38
h e a r t & l u n g 4 0 ( 2 0 1 1 ) e 3 2 ee 4 0
or PES) after the recanalization of CTOs is significantly superior to BMS in reducing the rates of long-term MACE and restenosis, and in reducing rates of TLR, TVR, and TVF. Implantations of DES appear to be more effective than of BMS for the treatment of CTO during long-term periods. The strategy of reopening a CTO lesion by restoring blood flow to a hibernating myocardium was demonstrated to improve symptoms and left ventricular function, though with relatively low initial success rates and high rates of restenosis.20,21 Stent implantation was one of the strategies to restore blood flow to CTO lesions. Until now, several studies demonstrated its efficiency in reopening CTO lesions, remarkably reducing the long-term stenosis rate.22-24 With its lower restenosis rate and better clinical outcomes compared with BMS, DES has been widely used in clinical practice for CTO lesions in recent years. However, most of the studies were nonrandomized controlled trials and of relatively small sample size. Therefore, compared with other subsets, a lack of extensive evidence remains when determining the efficiency of DES implantation in the treatment of CTO lesions. The results of this meta-analysis confirmed that the use of DES implantation for CTO lesions could significantly reduce long-term rates of MACE, regardless of whether follow-up periods were limited to 6 months or extended to >6 months (P < .00001). However, no difference was evident for in-hospital MACE rates between the 2 groups (P ¼ .86). Similar findings were observed in the analysis of MIs and all-cause death (P > .1). Compared with the BMS group, the rates of TLR, TVR, and TVF were significantly lower in the DES group (P < .00001). The main benefit of DES may lie in the reduced need for repeat revascularization procedures. Furthermore, the long-term rates of late lumen loss, stent restenosis, and reocclusions were significantly lower in the DES group than in the BMS group (P < .00001). The average long-term stent restenosis rate was 1.0% in the DES group and 39.7% in the BMS group. The mean long-term stent reocclusion rates were 4.3% and 11.5% in the DES and BMS groups, respectively. Dual antiplatelet therapy after PCI has been used since 1994, after the first BMS was approved.25 In 2006, the American College of Cardiology and American Heart Association guidelines recommended 1 month of aspirin and clopidogrel after the elective implantation of a BMS, 3 months after DES, and 6 months after PES.26 The Food and Drug Administration’s recommendations proposed dual antiplatelet therapy for 12 months after stenting in patients at a low risk of bleeding.27 In our meta-analysis, all selected patients received dual antiplatelet therapy with a thienopyridine (clopidogrel or ticlopidine) in combination with aspirin after stenting. Dual antiplatelet regimens were administered for at least 3 months or 6 months after DES implantation, and 1 month after BMS implantation, in 4 studies.10,12,16,19 In other studies,11,14,15,17,18 however, all patients received at least 6 months of aspirin and clopidogrel, after implantations of both BMS and DES.
Although 4 studies showed that the power of antiplatelet therapy in the DES group was stronger than in the BMS group, the long-term survival rate of patients receiving DES did not surpass that in the BMS group, consistent with results of Han et al.28 Moreover, we could not conclude whether the higher MACE-free survival rate of DES for CTO lesions was based on prolonged dual antiplatelet therapy. Although several studies demonstrated the benefits of DES implantation for the treatment of CTO lesions, most of them were nonrandomized controlled trials with relatively small sample sizes. Furthermore, the clinical follow-up periods in most studies usually ranged from 6 to 12 months. All of these limited the assessment of DES efficiency in the treatment of CTO lesions. As the first study comparing 5-year clinical outcomes of DES with BMS implantation for CTO lesions, Han et al28 also indicated the superiority of DES implantations in reducing the risks of TVR and MACE. However, no difference in survival rates was found between the 2 groups. Moreover, it enrolled a relatively larger sample size of 1184 patients. Hence, its clinical outcomes provided more compelling evidence for the long-term therapeutic effects of DES for CTO lesions (up to 5 years). Although it was excluded from our meta-analysis because of an absence of the primary and secondary endpoints we needed, its results were consistent with those of our meta-analysis, and further demonstrated the credibility of our conclusions. As with any meta-analysis, several limitations arise from using data in multiple studies of a similar nature. The aim of this analysis was to provide a comprehensive synthesis of studies on DES vs. BMS implantation for CTO lesions. By searching all related articles, we identified 10 studies, including 8 nonrandomized controlled trials and 2 randomized controlled trials. This was prone to exaggerate the treatment effect and decrease the intensity of the analysis. As listed in the selection criteria, studies that were deemed to lack sufficient data because of incomplete or absent baseline or follow-up data were excluded. This might increase the publication bias. Moreover, angiographic follow-up was not obtained in all patients, because only those individuals presenting evidence of ischemia underwent repeat angiography. Therefore, more large-scale, high-quality, prospective, randomized controlled trials are required to provide more reliable clinical and angiographic evidence that should reduce the variations introduced by the operator’s experience, operation settings, and number of patients included.
Conclusions The results of this meta-analysis confirm the longterm efficacy of DES over BMS for the treatment of CTOs. Implantations of DES could reduce the rates of stent restenosis, reocclusion, and MACE, as well as the need for repeated angiography.
h e a r t & l u n g 4 0 ( 2 0 1 1 ) e 3 2 ee 4 0
References
1. Stone GW, Kandzari DE, Mehran R, Colombo A, Schwartz RS, Bailey S, et al. Percutaneous recanalization of chronically occluded coronary arteries: a consensus document: part I. Circulation 2005;112:2364-72. 2. Sirnes PA, Myreng Y, Mølstad P, Bonarjee V, Golf S. Improvement in left ventricular ejection fraction and wall motion after successful recanalization of chronic coronary occlusions. Eur Heart J 1998;19:273-81. 3. Suero JA, Marso SP, Jones PG, Laster SB, Huber KC, Giorqi LV, et al. Procedural outcomes and long longterm survival among patients undergoing percutaneous coronary intervention of a chronic total occlusion in native coronary arteries: a 20-year experience. J Am Coll Cardiol 2001;38:409-14. 4. Stone GW, Colombo A, Teirstein PS, Moses JW, Leon MB, Reifart NJ, et al. Percutaneous recanalization of chronically occluded coronary arteries: procedural techniques, devices, and results. Catheter Cardiovasc Intervent 2005;66:217-36. 5. Moses JW, Leon MB, Popma JJ, Fitzgerald PJ, Holmes DR, O’Shaughnessy C, et al. Sirolimus-eluting stents versus standard stents in patients with stenosis in a native coronary artery. N Engl J Med 2003;349:1315-23. 6. Morice MC, Serruys PW, Sousa JE, Fajadet J, Ban Hayashi E, Perin M, et al. A randomized comparison of a sirolimus-eluting stent with a standard stent for coronary revascularization. N Engl J Med 2002;346: 1773-8. 7. Stone GW, Ellis SG, Cox DA, Hermiller J, O’Shaughnessy C, Mann JT, et al. A polymer-based, paclitaxel-eluting stent in patients with coronary artery disease. N Engl J Med 2004;350:221-31. 8. Higgins JPT, Green S, editors. Cochrane handbook for systematic reviews of interventions 4.2.6. The Cochrane Library, issue 4. Chichester, UK: John Wiley & Sons, Ltd; 2006. 9. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analysis. Br Med J [Clin Res] 2003;327:557e6. 10. Werner GS, Krack A, Schwarz G, Prochnau D, Betge S, Figulla HR. Prevention of lesion recurrence in chronic total coronary occlusions by paclitaxel-eluting stents. J Am Coll Cardiol 2004;44:2301-6. 11. Hoye A, Tanabe K, Lemos PA, Aoki J, Saia F, Arampatzis C, et al. Significant reduction in restenosis after the use of sirolimus-eluting stents in the treatment of chronic total occlusions. J Am Coll Cardiol 2004;43:1954-8. 12. Ge L, Iakovou I, Cosgrave J, Chieffo A, Montorfano M, Michev I, et al. Immediate and mid-term outcomes of sirolimus-eluting stent implantation for chronic total occlusions. Eur Heart J 2005;26:1056-62. 13. Nakamura S, Muthusamy TS, Bae JH, Cahyadi YH, Udayachalerm W, Tresukosol D. Impact of sirolimuseluting stent on the outcome of patients with chronic total occlusions. Am J Cardiol 2005;95:161-6. 14. Kelbaek H, Helqvist S, Thuesen L, Kløvgaard L, Jørgensen E, Saunama¨ki K, et al. Sirolimus versus bare metal stent implantation in patients with total coronary occlusions: subgroup analysis of the Stenting Coronary Arteries in Non-Stress/Benestent Disease (SCANDSTENT) Trial. Am Heart J 2006;152: 882-6.
e39
15. Suttorp MJ, Laarman GJ, Rahel BM, Kelder JC, Bosschaert MA, Kiemeneij F, et al. Primary Stenting of Totally Occluded Native Coronary Arteries II (PRISON II): a randomized comparison of bare metal stent implantation with sirolimus-eluting stent implantation for the treatment of total coronary occlusions. Circulation 2006;114:921-8. 16. Werner GS, Schwarz G, Prochnau D, Fritzenwanger M, Krack A, Betge S, et al. Paclitaxel-eluting stents for the treatment of chronic total coronary occlusions: a strategy of extensive lesion coverage with drugeluting stents. Cathet Cardiovasc Intervent 2006;67: 1-9. 17. Migliorini A, Moschi G, Vergara R, Parodi G, Carrabba N, Antoniucci D. Drug-eluting stentsupported percutaneous coronary intervention for chronic total coronary occlusion. Cathet Cardiovasc Intervent 2006;67:344-8. 18. Garcı´a-Garcı´a HM, Daemen J, Kukreja N, Tanimoto S, van Mieghem CA, van der Ent M, et al. Three-year clinical outcomes after coronary stenting of chronic total occlusion using sirolimus-eluting stents: insights from the Rapamycin-Eluting Stent Evaluated at Rotterdam Cardiology Hospital (RESEARCH) Registry. Cathet Cardiovasc Intervent 2007;70:635-9. 19. Felice F, Fiorilli R, Parma A, Menichelli M, Nazzaro MS, Pucci E, et al. Clinical outcome of patients with chronic total occlusion treated with drug-eluting stents. Int J Cardiol 2009;132:337-41. 20. Sirnes PA, Myreng Y, Mølstad P, Bonarjee V, Golf S. Improvement in left ventricular ejection fraction and wall motion after successful recanalization of chronic coronary occlusions. Eur Heart J 1998;19:273-81. 21. Rambaldi R, Hamburger JN, Geleijnse ML, Poldermans D, Kimman GJ, Aiazian AA, et al. Early recovery of wall motion abnormalities after recanalization of chronic totally occluded coronary arteries: a dobutamine echocardiographic, prospective, single center experience. Am Heart J 1998;136:831-6. 22. Rubartelli P, Niccoli L, Verna E, Giachero C, Zimarino M, Fontanelli A, et al. Stent implantation versus balloon angioplasty in chronic coronary occlusions: results from the GISSOC trial (Gruppo Italiano di Studio sullo Stent nelle Occlusioni Coronariche). J Am Coll Cardiol 1998;32:90-6. 23. Lotan C, Rozenman Y, Hendler A, Turgeman Y, Ayzenberg O, Beyar R, et al. Stents in total occlusion for restenosis prevention: the multicentre randomized STOP Study. The Israeli Working Group for Interventional Cardiology. Eur Heart J 2000;21: 1960-6. 24. Buller CE, Dzavik V, Carere RG, Mancini GB, Barbeau G, Lazzam C, et al. Primary stenting versus balloon angioplasty in occluded coronary arteries: the Total Occlusion Study of Canada (TOSCA). Circulation 1999;100:236-42. 25. Cross J. Antiplatelet therapy for improving post-PCI outcomes: interpreting current treatment guidelines for optimal management of the post-ACS patient. Am J Manag Care 2009;15(Suppl):S48-53. 26. Smith Jr SC, Feldman TE, Hirshfeld Jr JW, Jacobs AK, Kern MJ, King III SB, et al. ACC/AHA/SCAI 2005 guideline update for percutaneous coronary interventiondsummary article: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (ACC/ AHA/SCAI Writing Committee to Update the 2001
e40
h e a r t & l u n g 4 0 ( 2 0 1 1 ) e 3 2 ee 4 0
Guidelines for Percutaneous Coronary Intervention). J Am Coll Cardiol 2006;47:216-35. 27. Mayor S. Drug-eluting stents are safe for licensed indications, FDA panel says. Br Med J [Clin Res] 2006; 333:1235.
28. Han YL, Zhang J, Li Y, Wang SL, Jing QM, Yi XH, et al. Long-term outcomes of drug-eluting versus baremetal stent implantation in patients with chronic total coronary artery occlusions. Chin Med J [Engl] 2009;122:643-7.
Division of Cardiology, Shanghai Sixth People’s Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China b Division of Geriatrics, Shanghai Ren-Ji Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China c Department of Pathology, University of Western Ontario, London, Ontario, Canada d Critical Illness Research, Lawson Health Research Institute, Department of Medicine and Department of Pathology, University of Western Ontario, London, Ontario, Canada
article info
abstract
Article history: Received 21 May 2009 Revised 4 July 2010 Accepted 17 July 2010 Online 21 March 2011
Background: In the treatment of chronic total occlusions (CTOs), some uncertainty exists regarding the effect of drug-eluting stents (DESs) compared with the effects of bare mental stents (BMSs). We reviewed outcomes of DES vs. BMS implantation for CTO lesions, to evaluate the risk-benefit ratio of DES implantation.
Keywords: Drug-eluting stents Bare mental stents Chronic total occlusions Meta-analysis
Methods: Relevant studies of long-term clinical outcomes or angiographic outcomes of both BMS and DES implantation were examined. The primary endpoint comprised major adverse cardiovascular events (MACEs), including allcause deaths, myocardial infarctions (MIs), and target lesion revascularizations (TLRs). A fixed-effect model and random-effect model were used to analyze the pooling results. Results: Ten studies were included according to the selection criteria. Eight were nonrandomized controlled trials, and two consisted of a randomized controlled comparison between DES and BMS implantation. No significant difference was evident for in-hospital MACE rates between the two groups (odds ratio [OR], 1.07; 95% confidence interval [CI], .53 to 2.13), but the long-term MACE rates in the DES group were significantly lower than in the BMS group (OR, .22; 95% CI, .13 to .38; P < .00001). The rates of stent restenosis and reocclusions were also significantly lower in the DES group (OR, .14; 95% CI, .09 to .20; and OR, .23; 95% CI, .12 to .41, respectively). Conclusion: Implantation of the DES improves long-term angiographic and clinical outcomes compared with BMS in the treatment of CTO lesions. Cite this article: Ma, J., Yang, W., Singh, M., Peng, T., Fang, N., & Wei, M. (2011, MAY/JUNE). Meta-analysis of long-term outcomes of drug-eluting stent implantations for chronic total coronary occlusions. Heart & Lung, 40(3), e32-e40. doi:10.1016/j.hrtlng.2010.07.009.
* Corresponding author: Meng Wei, MD, PhD, Division of Cardiology, Shanghai Sixth People’s Hospital, No. 600 Yishan Road, Shanghai 200233, People’s Republic of China. E-mail address: [email protected] (M. Wei). y These authors equally contributed to this work 0147-9563/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.hrtlng.2010.07.009
h e a r t & l u n g 4 0 ( 2 0 1 1 ) e 3 2 ee 4 0
A chronic total coronary artery occlusion (CTO) is defined as the complete obstruction of coronary arteries with thrombolysis in myocardial infarction (TIMI) flow grade 0, with an estimated duration of >3 months. It is one of the most challenging lesion subsets in percutaneous coronary intervention (PCI). Although bare mental stents (BMSs) are widely applied clinically and, upon successful recanalization of CTO, were demonstrated to improve left ventricular function and survival,1-3 the major complications of high rates of restenosis and reocclusions limit its long-term usefulness.4,5 Drug-eluting stents (DESs), including sirolimus-eluting and paclitaxel-eluting varieties (SES and PES, respectively), are able to reduce the risk of restenosis.6,7 A few studies reported, with conflicting results, the clinical and angiographic outcomes of DESbased PCI for the treatment of CTO compared with BMS-based PCI. However, most of these were limited because they were nonrandomized controlled trials (RCTs) with a small sample size. The aim of our metaanalysis was to evaluate the risk-benefit balance of DES implantation vs. that of BMS implantation in the treatment of CTO.
Methods Search Strategy and Selection Criteria Two trained investigators (J.M. and W.Y.) independently searched Medline, EMbase, the Chinese Biomedical Database (from 1980 to April 2009), and the Cochrane Library (Issue 3, 2009). The reference lists of retrieved papers and reviews were also checked. No language restriction was used. The main search keywords included “chronic total occlusions,” “CTO,” “drugeluting stents,” “DES,” and “clinical trial*.” The inclusion criteria for retrieved studies comprised: (1) comparisons of DES implantation vs. BMS implantation, (2) patients with CTO diseases, and (3) a follow-up period of at least 6 months. Ongoing studies, duplicate publications, and those with incomplete data or a lack of baseline or follow-up data were excluded from this analysis.
Data Extraction and Study Characteristics Two investigators (J.M. and W.Y.) independently selected trials and extracted data on prespecified data forms, with disagreements resolved by discussion between them or, when necessary, in consultation with an arbiter (M.W). The primary endpoint comprised major adverse cardiovascular events (MACEs). The secondary end points included all-cause death, myocardial infarction (MI), target vessel revascularization (TVR), target lesion revascularization (TLR), target vessel failure (TVF), and * The asterisk is a truncated symbol, indicating to search this word by right truncation.
e33
angiographic outcomes involving minimal lumen diameter (MLD), late lumen loss, restenosis, and reocclusion. The MACEs were a composite of all-cause death, MIs, and TLRs. A TLR was defined as an ischemiadriven percutaneous or surgical revascularization of the target lesion attributable to restenosis within the stent, or 5 mm distal or proximal to the stent, after the initial procedure. A TVR was defined as a repeated revascularization within the treated vessel. Target vessel failures were defined as a composite of deaths from cardiac causes, MIs, and ischemia-driven TVRs. Late lumen loss was determined as the change in MLD from immediately after stent implantation to followup. Restenosis was defined as a diameter stenosis of 50% both within the stent and in the adjacent segments 5 mm proximal and distal, as determined by follow-up angiography. Reocclusion was defined as a recurrent total occlusion at the previous angioplasty site.
Quality Assessment Two investigators (W.Y. and J.M) independently assessed the quality of eligible articles according to the established methods of the Cochrane Handbook,8 which includes 5 items: randomization, allocation concealment, blinding, reported loss to follow-up, or withdrawal and comparability of baseline. Disagreements were resolved by discussions between them and, if necessary, by consultation with an arbiter (M.W.).
Statistical Analysis Statistical analysis was performed using Review Manager 5.0.18 (Cochrane Collaboration, Oxford, UK). Dichotomous variables were reported as odds ratios (ORs) and 95% confidence interval (CIs), whereas continuous variables were reported as weighted mean differences (WMDs) and 95% CIs. The c2 test and I2 values were calculated as a measure of statistical heterogeneity.8 If the I2 value was 0%, it denoted no observed statistical heterogeneity among studies, whereas a greater I2 value denoted more increasing heterogeneity.9 If no statistical heterogeneity was evident, dichotomous and continuous variables were compared using a fixed-effects model. If statistical heterogeneity was evident among studies, we first found the potential reasons, and then performed a sensitivity analysis or subgroup analysis, or analyzed studies using a random-effects model. A funnel plot was performed to assess the presence of publication bias.
Results The initial search generated 243 reports, of which 233 were excluded because of inconsistencies with
e34
Table 1 e Characteristics of 10 included studies Study
DM (%)
LVEF (%)
HBP (%)
Previous MI (%)
Hyperlipidemia (%)
96
64 9
77
31
68 16
79
45
76
AþC
Hoye et al (2004)11 Lei et al (2005)12
84 381
60 11 61 10
76 87
12 22
UN 53 11
39 45
52 61
56 49
AþC A þ T/C
Nakamura et al (2005)13 Kelbaek et al (2006)14 Suttorp (2006)15
180
69 8
65
33
54 7
59
36
34
AþC
127 200
62 10 60 10
79 80
20 13
51 11 UN
41 46
60 49
89 90
AþC AþC
Werner et al (2006)16
122
63 10
78
33
64 17
77
45
81
AþC
Migliorini et al (2006)17 Hecto et al (2007)18
118
66 11
UN
19
UN
53
50
47
AþC
147
61 11
71
10
UN
39
51
62
AþC
Felice et al (2008)19
223
62 10
76
31
51 10
71
67
66
A þ T/C
Werner et al (2004)10
Oral antiplatelet regimen
Duration of antiplatelet regimen DES 6 m; BMS 1m 6m T/C 3 m (DES), 1 m (BMS), A maintained UN 12 m C 6 m; A indefinite DES 6 m; BMS 1 m C 6 m; A indefinite A þ C 6 m; A lifelong T/C 6 m (DES), 1 m (BMS); A indefinite
SES/ PES (%)
Clinical followup
Angiographic follow-up
0/100
12 m
6m
100/0 100/0
12 m 6m
6m 6-8 m
100/0
12 m
6-12 m
100/0 100/0
7m 6-12 m
6m 6m
0/100
12 m
6m
UN
6m
6m
100/0
36 m
36 m
73/38
18 m
18 m
DM, diabetes mellitus; LVEF, left-ventricular ejection fraction; HBP, high blood pressure; MI, myocardial infarction; DES, drug-eluting stents; BMS, bare mental stents; SES, sirolimus-eluting stents; PES, paclitaxel-eluting stents; A, aspirin; C, clopidogrel; T, ticlopidine; UN, unclear; m, months; y, years.
h e a r t & l u n g 4 0 ( 2 0 1 1 ) e 3 2 ee 4 0
No. of Mean Males patients age (y) (%)
h e a r t & l u n g 4 0 ( 2 0 1 1 ) e 3 2 ee 4 0
Figure 1 e Funnel plot of drug-eluting stents vs. bare mental stents for chronic total occlusions.
selection criteria. Ultimately, 10 published studies were included,10-19 enrolling 1678 subjects undergoing a PCI with DES or BMS implantations. Two studies14,15 were RCTs, and 810-13,16-19 were nonrandomized controlled comparisons. All studies were published between 2004 and 2008 inclusive. Among these, 6 studies11-15,18 used SES for the treatment of CTOs, and 2 studies10,16 used PES for the treatment of CTOs, whereas the remainder17,19 used both SES and PES for the treatment of CTOs. The characteristics of the included studies are presented in Table 1. The mean age of patients was 63 years. On average, males accounted for 76.5% of subjects. The clinical and angiographic median follow-up periods were 13 months (range, 6 to 36 months) and 6 months (range, 6 to 12 months), respectively. The proportion of patients with diabetes mellitus ranged from 10% to 33% across trials. About 52% of the patients had experienced a previous MI. Approximately 55% and 65% of patients had a history of hypertension and hyperlipidemia, respectively. The average baseline left-ventricular ejection fraction (LVEF), as reported in six studies,10,12-14,16,19 was 57% 12% standard deviation (SD). Funnel plots did not suggest any evidence of publication bias (Figure 1).
Data Synthesis With respect to all endpoints, a fixed-effects model was used to analyze statistical heterogeneity among studies, and to combine data from enrolled studies. The results are shown in Figures 2 to 6 and Table 2.
e35
group (Figure 2). With regard to long-term MACE rates, however, significant statistical heterogeneity was evident among studies (P ¼ .002; I2, 65%). Inconsistent MACE definitions in different studies could have created the statistical heterogeneity. The different clinical settings in which procedures were performed also added heterogeneity to the analysis. Moreover, the number of enrolled patients varied widely among studies and between the 2 groups, which also made long-term MACE rates different. When reanalyzed using the random-effects model, the long-term MACE rates were significantly lower in the DES group than in the BMS group (OR, .22; 95% CI, .13 to .38; P < .00001) (Figure 3). To eliminate the statistical heterogeneity, we performed a subgroup analysis of the primary endpoint according to different follow-up periods. In studies limiting follow-up periods to 6 months,12,15,17 no statistical heterogeneity was evident (P ¼ .57; I2, 0%), but the long-term MACE rates were significantly lower in the DES-based PCI group than in the BMSbased PCI group (OR, .35; 95% CI, .23 to .54; P < .00001). In studies extending follow-up periods to >6 months,10,11,13-16,18,19 statistical heterogeneity was still present (P ¼ .003; I2, 68%). Finally, sensitivity analyses were conducted after the exclusion of 2 studies.11,18 No statistical heterogeneity was evident among studies, and the DES group had significantly lower long-term MACE rates than the BMS group (OR, .13; 95% CI, .09 to .20; P < .00001).
MI Data on the rate of in-hospital and long-term MIs were available in 5 studies10,12,13,16,19 and 7 studies,11-15, 18, 19 respectively. As depicted in Figures 4 and 5, no statistical heterogeneity was evident among studies (P ¼ .89; I2, 0%; and P ¼ .94; I2, 0%, respectively), with no significant difference between the DES and BMS groups (OR, 1.14; 95% CI, .57 to 2.29; P ¼ .72; and OR, .84; 95% CI, .46 to 1.54; P ¼ .58, respectively) (Figures 4, 5).
All-cause Death The rate of long-term mortality, including cardiac and noncardiac death, was reported in 7 studies.10,12,15-19 As shown in Figure 6, no statistical heterogeneity was evident among studies (P ¼ .80; I2, 0%), with no significant difference between the DES and BMS groups (OR, 1.02; 95% CI, .49 to 2.10; P ¼ .97) (Figure 6).
TLR
Clinical Endpoints MACE Five studies10,12,13,16,19 reported both in-hospital and long-term rates of MACE, whereas the remainder11,14,15,17,18 only described long-term MACE rates. A subanalysis was performed according to different observation times. Regarding in-hospital MACE rates, no statistical heterogeneity (P ¼ .93; I2, 0%) and no significant difference (OR, 1.07; 95% CI, .53 to 2.13; P ¼ .86) were present between the DES group and BMS
Data on long-term TLR were available in 8 studies.10,12-16,18,19 Statistical heterogeneity was evident among these studies (P ¼ .05; I2, 51%). Sensitivity analyses, conducted after the exclusion of 1 study,18 showed no statistical heterogeneity (P ¼ .29; I2, 18%) but a significantly lower rate of TLR in the DES group vs. the BMS group (OR, .13; 95% CI, .08 to .19, P < .00001).
TVR Regarding the endpoint of long-term TVR, data were available in 7 studies.11,12,14,15,17-19 No statistical
e36
h e a r t & l u n g 4 0 ( 2 0 1 1 ) e 3 2 ee 4 0
Figure 2 e Pooled results of in-hospital major adverse cardiovascular events in patients undergoing percutaneous drug-eluting stent (DES) placement for chronic total occlusions. BMS, bare mental stents, M-H, Mantel-Haenszel.
Figure 3 e Pooled results of long-term MACEs in patients undergoing percutaneous drug-eluting stent (DES) placement for chronic total occlusions. BMS, bare mental stents, M-H, Mantel-Haenszel.
Figure 4 e Pooled results of in-hospital myocardial infarctions in patients undergoing percutaneous drugeluting stent (DES) placement for chronic total occlusions. BMS, bare mental stents, M-H, Mantel-Haenszel.
heterogeneity was evident among studies (P ¼ .29; I2, 18%), but the long-term TVR rate in the DES group was significantly lower than in the BMS group (OR, .25, 95% CI, .17 to .36; P < .00001).
TVF Three studies10,14,15 reported the long-term rates of TVF. No statistical heterogeneity (P ¼ .32; I2, 11%) was found among studies. Compared with the BMS group, the rate of long-term TVF was significantly lower in the DES group (OR, .16; 95% CI, .09 to .29; P < .00001).
Angiographic Outcomes The long-term coronary angiography analysis comprised 10 studies of 479 patients in the DES group, and 604 patients in the BMS group. The average angiographic follow-up time was 6 months. With regard to all angiographic endpoints except for late lumen loss, no statistical heterogeneity was evident among studies. Compared with the BMS group, the rate of MLD was significantly higher in the DES group (WMD, 1.00; 95% CI, .91 to 1.09; P < .00001), but the rate
e37
h e a r t & l u n g 4 0 ( 2 0 1 1 ) e 3 2 ee 4 0
Figure 5 e Pooled results of long-term myocardial infarctions in patients undergoing percutaneous drugeluting stent (DES) placement for chronic total occlusions. BMS, bare mental stents, M-H, Mantel-Haenszel.
Figure 6 e Pooled results of long-term death due to any cause in patients undergoing percutaneous drugeluting stent (DES) placement for chronic total occlusions. BMS, bare mental stents, M-H, Mantel-Haenszel.
Table 2 e Summary effects for subanalysis of MACEs, TLR, TVR, TVF, and angiographic endpoints Outcomes Subanalysis of MACEs Limiting follow-up to 6 months Extending follow-up to >6 months TLR TVR TVF Angiographic endpoints MLD Late lumen loss Restenosis Reocclusion
OR (95% CI) .35 .13 .13 .25 .16
WMD (95% CI)
(.23 to .54) (.09 to .20) (.08 to .19) (.17 to .36) (.09 to .29) 1.00 (.91 to 1.09) 1.04 (1.21 to .07)
.14 (.09 to .20) .23 (.12 to .41)
P
I2 (%)
.57 .11 .29 .29 .32
0 45.0 17.8 18.0 11.4
.17 .07 .24 .43
34.9 53.2 24.8 0
MACEs, major adverse cardiovascular events; TLR, target lesion revascularization; TVR, target vessel revascularization; TVF, target vessel failure; MLD, minimal lumen diameter; OR, risk ratio; WMD, weighted mean difference; 95% CI, 95% confidence interval.
of late lumen loss was significantly lower in the DES group (WMD, 1.04; 95% CI, 1.21 to .87; P < .00001). Furthermore, the rates of stent restenosis and reocclusion in the DES group were significantly lower than in the BMS group (WMD, .14; 95% CI, .09 to .20; P < .00001; and WMD, .23; 95% CI, .12 to .41, P < .00001, respectively).
Discussion The present meta-analysis focused on the clinical and angiographic efficacy of DES implantation for the treatment of CTO lesions compared with BMS implantation. The results indicate that DES (either SES
e38
h e a r t & l u n g 4 0 ( 2 0 1 1 ) e 3 2 ee 4 0
or PES) after the recanalization of CTOs is significantly superior to BMS in reducing the rates of long-term MACE and restenosis, and in reducing rates of TLR, TVR, and TVF. Implantations of DES appear to be more effective than of BMS for the treatment of CTO during long-term periods. The strategy of reopening a CTO lesion by restoring blood flow to a hibernating myocardium was demonstrated to improve symptoms and left ventricular function, though with relatively low initial success rates and high rates of restenosis.20,21 Stent implantation was one of the strategies to restore blood flow to CTO lesions. Until now, several studies demonstrated its efficiency in reopening CTO lesions, remarkably reducing the long-term stenosis rate.22-24 With its lower restenosis rate and better clinical outcomes compared with BMS, DES has been widely used in clinical practice for CTO lesions in recent years. However, most of the studies were nonrandomized controlled trials and of relatively small sample size. Therefore, compared with other subsets, a lack of extensive evidence remains when determining the efficiency of DES implantation in the treatment of CTO lesions. The results of this meta-analysis confirmed that the use of DES implantation for CTO lesions could significantly reduce long-term rates of MACE, regardless of whether follow-up periods were limited to 6 months or extended to >6 months (P < .00001). However, no difference was evident for in-hospital MACE rates between the 2 groups (P ¼ .86). Similar findings were observed in the analysis of MIs and all-cause death (P > .1). Compared with the BMS group, the rates of TLR, TVR, and TVF were significantly lower in the DES group (P < .00001). The main benefit of DES may lie in the reduced need for repeat revascularization procedures. Furthermore, the long-term rates of late lumen loss, stent restenosis, and reocclusions were significantly lower in the DES group than in the BMS group (P < .00001). The average long-term stent restenosis rate was 1.0% in the DES group and 39.7% in the BMS group. The mean long-term stent reocclusion rates were 4.3% and 11.5% in the DES and BMS groups, respectively. Dual antiplatelet therapy after PCI has been used since 1994, after the first BMS was approved.25 In 2006, the American College of Cardiology and American Heart Association guidelines recommended 1 month of aspirin and clopidogrel after the elective implantation of a BMS, 3 months after DES, and 6 months after PES.26 The Food and Drug Administration’s recommendations proposed dual antiplatelet therapy for 12 months after stenting in patients at a low risk of bleeding.27 In our meta-analysis, all selected patients received dual antiplatelet therapy with a thienopyridine (clopidogrel or ticlopidine) in combination with aspirin after stenting. Dual antiplatelet regimens were administered for at least 3 months or 6 months after DES implantation, and 1 month after BMS implantation, in 4 studies.10,12,16,19 In other studies,11,14,15,17,18 however, all patients received at least 6 months of aspirin and clopidogrel, after implantations of both BMS and DES.
Although 4 studies showed that the power of antiplatelet therapy in the DES group was stronger than in the BMS group, the long-term survival rate of patients receiving DES did not surpass that in the BMS group, consistent with results of Han et al.28 Moreover, we could not conclude whether the higher MACE-free survival rate of DES for CTO lesions was based on prolonged dual antiplatelet therapy. Although several studies demonstrated the benefits of DES implantation for the treatment of CTO lesions, most of them were nonrandomized controlled trials with relatively small sample sizes. Furthermore, the clinical follow-up periods in most studies usually ranged from 6 to 12 months. All of these limited the assessment of DES efficiency in the treatment of CTO lesions. As the first study comparing 5-year clinical outcomes of DES with BMS implantation for CTO lesions, Han et al28 also indicated the superiority of DES implantations in reducing the risks of TVR and MACE. However, no difference in survival rates was found between the 2 groups. Moreover, it enrolled a relatively larger sample size of 1184 patients. Hence, its clinical outcomes provided more compelling evidence for the long-term therapeutic effects of DES for CTO lesions (up to 5 years). Although it was excluded from our meta-analysis because of an absence of the primary and secondary endpoints we needed, its results were consistent with those of our meta-analysis, and further demonstrated the credibility of our conclusions. As with any meta-analysis, several limitations arise from using data in multiple studies of a similar nature. The aim of this analysis was to provide a comprehensive synthesis of studies on DES vs. BMS implantation for CTO lesions. By searching all related articles, we identified 10 studies, including 8 nonrandomized controlled trials and 2 randomized controlled trials. This was prone to exaggerate the treatment effect and decrease the intensity of the analysis. As listed in the selection criteria, studies that were deemed to lack sufficient data because of incomplete or absent baseline or follow-up data were excluded. This might increase the publication bias. Moreover, angiographic follow-up was not obtained in all patients, because only those individuals presenting evidence of ischemia underwent repeat angiography. Therefore, more large-scale, high-quality, prospective, randomized controlled trials are required to provide more reliable clinical and angiographic evidence that should reduce the variations introduced by the operator’s experience, operation settings, and number of patients included.
Conclusions The results of this meta-analysis confirm the longterm efficacy of DES over BMS for the treatment of CTOs. Implantations of DES could reduce the rates of stent restenosis, reocclusion, and MACE, as well as the need for repeated angiography.
h e a r t & l u n g 4 0 ( 2 0 1 1 ) e 3 2 ee 4 0
References
1. Stone GW, Kandzari DE, Mehran R, Colombo A, Schwartz RS, Bailey S, et al. Percutaneous recanalization of chronically occluded coronary arteries: a consensus document: part I. Circulation 2005;112:2364-72. 2. Sirnes PA, Myreng Y, Mølstad P, Bonarjee V, Golf S. Improvement in left ventricular ejection fraction and wall motion after successful recanalization of chronic coronary occlusions. Eur Heart J 1998;19:273-81. 3. Suero JA, Marso SP, Jones PG, Laster SB, Huber KC, Giorqi LV, et al. Procedural outcomes and long longterm survival among patients undergoing percutaneous coronary intervention of a chronic total occlusion in native coronary arteries: a 20-year experience. J Am Coll Cardiol 2001;38:409-14. 4. Stone GW, Colombo A, Teirstein PS, Moses JW, Leon MB, Reifart NJ, et al. Percutaneous recanalization of chronically occluded coronary arteries: procedural techniques, devices, and results. Catheter Cardiovasc Intervent 2005;66:217-36. 5. Moses JW, Leon MB, Popma JJ, Fitzgerald PJ, Holmes DR, O’Shaughnessy C, et al. Sirolimus-eluting stents versus standard stents in patients with stenosis in a native coronary artery. N Engl J Med 2003;349:1315-23. 6. Morice MC, Serruys PW, Sousa JE, Fajadet J, Ban Hayashi E, Perin M, et al. A randomized comparison of a sirolimus-eluting stent with a standard stent for coronary revascularization. N Engl J Med 2002;346: 1773-8. 7. Stone GW, Ellis SG, Cox DA, Hermiller J, O’Shaughnessy C, Mann JT, et al. A polymer-based, paclitaxel-eluting stent in patients with coronary artery disease. N Engl J Med 2004;350:221-31. 8. Higgins JPT, Green S, editors. Cochrane handbook for systematic reviews of interventions 4.2.6. The Cochrane Library, issue 4. Chichester, UK: John Wiley & Sons, Ltd; 2006. 9. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analysis. Br Med J [Clin Res] 2003;327:557e6. 10. Werner GS, Krack A, Schwarz G, Prochnau D, Betge S, Figulla HR. Prevention of lesion recurrence in chronic total coronary occlusions by paclitaxel-eluting stents. J Am Coll Cardiol 2004;44:2301-6. 11. Hoye A, Tanabe K, Lemos PA, Aoki J, Saia F, Arampatzis C, et al. Significant reduction in restenosis after the use of sirolimus-eluting stents in the treatment of chronic total occlusions. J Am Coll Cardiol 2004;43:1954-8. 12. Ge L, Iakovou I, Cosgrave J, Chieffo A, Montorfano M, Michev I, et al. Immediate and mid-term outcomes of sirolimus-eluting stent implantation for chronic total occlusions. Eur Heart J 2005;26:1056-62. 13. Nakamura S, Muthusamy TS, Bae JH, Cahyadi YH, Udayachalerm W, Tresukosol D. Impact of sirolimuseluting stent on the outcome of patients with chronic total occlusions. Am J Cardiol 2005;95:161-6. 14. Kelbaek H, Helqvist S, Thuesen L, Kløvgaard L, Jørgensen E, Saunama¨ki K, et al. Sirolimus versus bare metal stent implantation in patients with total coronary occlusions: subgroup analysis of the Stenting Coronary Arteries in Non-Stress/Benestent Disease (SCANDSTENT) Trial. Am Heart J 2006;152: 882-6.
e39
15. Suttorp MJ, Laarman GJ, Rahel BM, Kelder JC, Bosschaert MA, Kiemeneij F, et al. Primary Stenting of Totally Occluded Native Coronary Arteries II (PRISON II): a randomized comparison of bare metal stent implantation with sirolimus-eluting stent implantation for the treatment of total coronary occlusions. Circulation 2006;114:921-8. 16. Werner GS, Schwarz G, Prochnau D, Fritzenwanger M, Krack A, Betge S, et al. Paclitaxel-eluting stents for the treatment of chronic total coronary occlusions: a strategy of extensive lesion coverage with drugeluting stents. Cathet Cardiovasc Intervent 2006;67: 1-9. 17. Migliorini A, Moschi G, Vergara R, Parodi G, Carrabba N, Antoniucci D. Drug-eluting stentsupported percutaneous coronary intervention for chronic total coronary occlusion. Cathet Cardiovasc Intervent 2006;67:344-8. 18. Garcı´a-Garcı´a HM, Daemen J, Kukreja N, Tanimoto S, van Mieghem CA, van der Ent M, et al. Three-year clinical outcomes after coronary stenting of chronic total occlusion using sirolimus-eluting stents: insights from the Rapamycin-Eluting Stent Evaluated at Rotterdam Cardiology Hospital (RESEARCH) Registry. Cathet Cardiovasc Intervent 2007;70:635-9. 19. Felice F, Fiorilli R, Parma A, Menichelli M, Nazzaro MS, Pucci E, et al. Clinical outcome of patients with chronic total occlusion treated with drug-eluting stents. Int J Cardiol 2009;132:337-41. 20. Sirnes PA, Myreng Y, Mølstad P, Bonarjee V, Golf S. Improvement in left ventricular ejection fraction and wall motion after successful recanalization of chronic coronary occlusions. Eur Heart J 1998;19:273-81. 21. Rambaldi R, Hamburger JN, Geleijnse ML, Poldermans D, Kimman GJ, Aiazian AA, et al. Early recovery of wall motion abnormalities after recanalization of chronic totally occluded coronary arteries: a dobutamine echocardiographic, prospective, single center experience. Am Heart J 1998;136:831-6. 22. Rubartelli P, Niccoli L, Verna E, Giachero C, Zimarino M, Fontanelli A, et al. Stent implantation versus balloon angioplasty in chronic coronary occlusions: results from the GISSOC trial (Gruppo Italiano di Studio sullo Stent nelle Occlusioni Coronariche). J Am Coll Cardiol 1998;32:90-6. 23. Lotan C, Rozenman Y, Hendler A, Turgeman Y, Ayzenberg O, Beyar R, et al. Stents in total occlusion for restenosis prevention: the multicentre randomized STOP Study. The Israeli Working Group for Interventional Cardiology. Eur Heart J 2000;21: 1960-6. 24. Buller CE, Dzavik V, Carere RG, Mancini GB, Barbeau G, Lazzam C, et al. Primary stenting versus balloon angioplasty in occluded coronary arteries: the Total Occlusion Study of Canada (TOSCA). Circulation 1999;100:236-42. 25. Cross J. Antiplatelet therapy for improving post-PCI outcomes: interpreting current treatment guidelines for optimal management of the post-ACS patient. Am J Manag Care 2009;15(Suppl):S48-53. 26. Smith Jr SC, Feldman TE, Hirshfeld Jr JW, Jacobs AK, Kern MJ, King III SB, et al. ACC/AHA/SCAI 2005 guideline update for percutaneous coronary interventiondsummary article: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (ACC/ AHA/SCAI Writing Committee to Update the 2001
e40
h e a r t & l u n g 4 0 ( 2 0 1 1 ) e 3 2 ee 4 0
Guidelines for Percutaneous Coronary Intervention). J Am Coll Cardiol 2006;47:216-35. 27. Mayor S. Drug-eluting stents are safe for licensed indications, FDA panel says. Br Med J [Clin Res] 2006; 333:1235.
28. Han YL, Zhang J, Li Y, Wang SL, Jing QM, Yi XH, et al. Long-term outcomes of drug-eluting versus baremetal stent implantation in patients with chronic total coronary artery occlusions. Chin Med J [Engl] 2009;122:643-7.